INTERVIEW WITH JONATHAN SCHWABISH

Jon Schwabish (USA) is a Senior Research Associate at The Urban Institute’s Income and Benefits Policy Center and founder of PolicyViz, a website to help people communicate research, analysis and ideas in a better way. Jon Schwabish: ‘’At this point, it may be safe to say that everyone knows they need data visualization, but it’s probably less true that not everyone knows they need to create good and effective data visualizations’’.

Graphic Hunters: What is the power of a good visualization? In what way can a visualization help to understand or to communicate information?
Jon Schwabish: People are generally more likely to remember and recognize information when it is presented to them visually. That’s why taking numbers out of dense tables and buried in text and placing them into graphs and other visualizations can help people gain insights and help them remember the information they’ve seen. Visualizations can also tap into people’s emotions and encourage them to connect with content in visceral ways.

I have read that someone has to understand a visualization within seconds, others say visualizations should leave something to explore. What is your opinion on this?
It often depends on the audience and context. On the one hand, you probably want the graph you share on Twitter, for example, to be easily recognized and understood in a moment. On the other hand, the interactive dashboard you are using with your colleagues may require more in-depth view and exploration. There are then graphs that sit somewhere in the middle – visualizations that give some people an immediate conclusion, but also leave space for others to explore and gain more insight and understanding.

What is the first most important question one should ask before starting visualizing the data?
What are you trying to communicate and to whom? Okay, that’s really two important questions, but the audience is so important to creating a graph and effectively communicating its contents, that we can’t forget them.

What tools are important to learn when you want to work with visualizations?  
It depends on what you are trying to create. If you are creating interactive visualizations, Excel is not going to get you very far. For online interactive visualizations, JavaScript is the programming language you will need to use, especially the d3 library. There are a variety of tools that have tried to make access to d3 easier and to remove some of the coding requirements; tools such as HighCharts and Lyra (for more dashboarding-type visualizations, tools like PowerBI and Tableau are popular). For static visualizations, Excel is my go-to tool and one that everybody knows and can learn fairly easily. But I typically start with paper and pencil – start analog, plan your visualization as best you can, and then go to the computer.

What is the role of design in a visualization?
Crucial. How do you get people to want to see your visualization, and then how to do keep them there for further exploration? Design elements such as color, font, layout, and user interface design all help in that effort. Creating effective visualizations is not about “making things pretty,” but about recognizing how to communicate and understanding how conscious – and oftentimes simple – design choices can help you do so. This is something I discuss a lot in my new book on giving better presentations.

What’s the biggest mistake often made in visualizations?
The biggest mistake is probably using the wrong graph for the data. There’s not a one-to-one mapping from data types to graph types (unfortunately), but some of the worst graphs try to cram some kind of data into a chart type that is clearly meant for a different data type; for example, showing time series data in a pie chart.

Who needs data visualization most and doesn’t know it?
At this point, it may be safe to say that everyone knows they need data visualization, but it’s probably less true that not everyone knows they need to create good and effective data visualizations. All sorts of people who work with data often fail to recognize that their reader or user hasn’t seen these data before and pack a graph full of data markers and labels and gridlines and tick marks and colors and…. It only makes the reader’s learning process entirely too difficult. Even with some basic strategies and education, I believe everyone can create effective visualizations.

If you could choose one visualization that you can put in a frame and hang in your living room, which visualization would you choose and why?
Actually, the one I’m looking at right now is my Graphic Continuum poster, a collection of nearly 90 different graphic types. Admittedly, I’m a little biased! I think anything created by Accurat Studios is probably worthy of putting on a wall, and I love some of the work from Hyperakt Studios as well.

What is the most frequently asked question in your training sessions and what is your answer to this?
Lately, I’ve been hearing a lot of questions about which data visualization tool people think they should learn. When I dig a little further, I find that the tools they are most interested in learning are not the tools that will help them communicate to their audience. For example, they want to learn how to create immersive interactive visualizations when their audience just wants printouts. There’s no need to train yourselves or others – or hire new people -to obtain skills that aren’t going to help you or your organization achieve your goals.

What lessons do you want the participants to take home for the training?
My goal for the training is for participants to understand the basics of effective visualizations and to always think about their audience. Define to whom create immersive interactive visualizations when their audience just wants printouts. There’s no need to train yourselves or others – or hire new people -to obtain skills that aren’t going to help you or your organization achieve your goals.